Populations

CHAPTER 3 Populations




Population: 1. The total number of persons inhabiting a country, city or any district. 2. The body of inhabitants of a place. 3. The assemblage of organisms under consideration. 4. An aggregate of statistical items.


—The American College Dictionary1


Statisticians think in terms of large numbers. They focus on the multitude rather than the individual. Once you make this transition, you will be on your way to understanding the theory behind biostatistics and the process of inference.



THEORETICAL CONCEPTS OF POPULATIONS


Most of us deal almost exclusively with individual situations throughout the day. The recommendations we make to patients depend upon the particulars of their condition. Through the formal education process, we have been trained to respond in a certain way in a given situation. Our actions are further influenced by our personal experience and some of us have accumulated quite an extensive and valuable collection.


The theory behind biostatistics expands on this principle. Imagine being able to observe the response to a given treatment (through a totally unbiased approach) of not just one patient, but of an infinite number of patients. Not all patients would have the same outcome. However, the results would tend to cluster and, if the treatment helped, the outcome would be better overall than with a different pathway that did not work as well. Armed with this extensive knowledge, you could make a prediction for an individual result based on the responses observed in the larger group.


We usually think of populations in terms of people who occupy a certain piece of land but, in statistical language, the definition is actually much broader. When we encounter the medical literature, the populations studied are usually groups of people with common, quantifiable characteristics. (Some statistics textbooks define populations as collections of data or observations. This refers to the actual data set taken from the collection of units or things being studied. For our purposes, we will use the more universal definition of populations as collections of the units themselves.)



• A population is a collection of things having some quantifiable characteristic in common.


We identify populations based on what we would like to study. A population could be composed of just about anything. It could be wine grapes harvested from a certain county which we would like to taste-test. It could be springs produced in a factory, of which we would like to test the tensile strength. It could be insect eggs or raindrops. It could be children with asthma or women with osteoporosis.


A true population is a theoretical concept. In theory, many populations cannot be counted completely since the number in the collection is extremely large. Take the example of raindrops. If we consider not only all raindrops that are falling at the current time but also those that have fallen in the past, and those yet to come, the true number approaches infinity. Many populations are so huge that it is impossible to account for each member, such as patients with congestive heart failure. A Venn diagram of a large population might have indistinct, muted edges or arrows pointing concentrically away from the center, as in Figure 3-1.


image

FIGURE 3-1 Venn diagrams of large populations with indistinct, muted edges and arrows pointing concentrically away from the center.


It is widely accepted to use a distinct circle to represent a population. When you see a traditional circle that represents a large population, try to think in terms of these concepts.


Another reason why populations do not have crisp edges is that, in many cases, they are constantly changing. Think of the population of kids with asthma. The definition of asthma is firm. In other words, the members of the population share a measurable characteristic, such as a collection of recurrent symptoms or the results of a breathing test. This is how we define the population. However, the members of the population are difficult to count at any given time. We set up criteria to distinguish members of the population, but new members are meeting these criteria every day while others are being disqualified. For example, new diagnoses of asthma are continually being made, which adds to the population. On the other hand, because children grow, there will be a continuous flux out of the population at whatever age (in minutes, perhaps seconds?) a “child” becomes “not a child.”



POPULATION PARAMETERS


Every population has attributes we attempt to measure. These are not the same as the characteristics that define the population, although the characteristics and attributes may overlap. The attributes are referred to as parameters. Even though populations are fluid, their attributes are quite stable. For instance, consider the parameter of “average age” in the population of U.S. citizens. If you were able to measure the average age of the population of the United States, it would not vary from day to day, even though many people enter this population through birth and immigration, and many exit through death and change of citizenship. The flux of people entering and exiting the population has a negligible effect on average age. Although populations in themselves can be immeasurable entities, they are very real and their parameters are unyielding. There is an average age of the U.S. population and it does not vary over days or even several weeks, although it may shift gradually over years.


A parameter may be something simple to grasp, like average age. It could also be something more abstract, like the risk of an accident in a population of drunken drivers or the survival rate in a population of people with a certain type of cancer. A population parameter is numerically very solid. We may not be able to know it exactly but we can estimate it with a degree of certainty, using statistical techniques.


The fact that the population attributes or parameters are firm is a very important concept in biostatistics. Consider the parameter of risk of an accident in a population of drunken drivers. We could define the population by the common characteristic of a blood alcohol level greater than a certain cutoff point. The population is further defined by those meeting the above blood alcohol criterion who are behind the wheel of a moving vehicle. You may recognize this as a population that is constantly changing. Individuals enter and exit this population all the time. There is no way to account for all of them at any one point.


The parameter of risk is a number that has a concrete value. It is the chance of an impaired driver having an accident within a unit of time when he or she is on the road. The risk exists and it is a fixed value, even though the population is fluid (in more ways than one!). We cannot exactly measure this risk but we can estimate it using observations. The point being made here is that any parameter of a population is stable and, although we may not have access to the exact number, we have the methods to estimate it using observations.



• A population parameter is a numerical value that summarizes the population data.

Related posts:

  1. Types of Statistical Tests
  2. Ethics in Medical Research
  3. Measures of Disease
  4. Mathematical Principles

Stay updated, free articles. Join our Telegram channel
Tags:
Jun 18, 2016 | Posted by in BIOCHEMISTRY | Comments Off on Populations

Full access? Get Clinical Tree
Get Clinical Tree app for offline access